ImmTACs: Bi-specific TCR-anti-CD3 fusions for potent redirected killing of cancer cells

Transcription

1 ImmTACs: Bi-specific TCR-anti-CD3 fusions for potent redirected killing of cancer cells Dr. Joseph Dukes, Head of Preclinical Biology, Immunocore Ltd 12 th November

2 The company - Immunocore Immunocore is based near Oxford, UK Immunocore is developing a unique biologic platform Creating soluble T-cell receptors ( monoclonal TCRs ) TCR targets specific peptide in context of Class I HLA Anti-cancer mtcr s engineered to produce bispecific drugs that modulate T-cell activity: Immune mobilising monoclonal TCRs Against Cancer (ImmTACs) Lead internal programme IMCgp100 in phase IIa in UK and USA Targeting malignant melanoma Strategic partnerships with GSK, Genentech, Medimmune and Eli Lilly (codevelopment) >115 staff 2

3 Antibody therapies have been the standard biologic approach in oncology for the last 15+ years Antibodies ADCs Bispecifics Cancer cell Cell surface protein CAR T cell and a range of approaches have been adopted, some with great success. However: Antibodies can target ~10% of available cancer targets (surface proteins) Number of target epitopes needs to be high ( ,000+ per cell) Distance matters (for effector function) variable with antibodies 3

4 The power of the immune system to rid its host of cancerous cells is being realised Cancer Immunotherapy: Science magazine s breakthrough of the year (2013) T-cells can prevent cancer Data shows that after 10 years immunosuppressed organ transplant recipients are 5-15 times more likely to be diagnosed with cancer Checkpoint blockage: Ipilimumab (anti-ctla4) and Nivolumab (anti-pd1) show durable responses that can be curative for cancer Relieving the cancer-induced suppression of T-cells allows these powerful components of the immune system to correctly recognise and eliminate cancer cells from the body 4

5 The basics of antigen presentation by Class I MHC (HLA) cancer cell antigen presented on cell surface peptides TAP proteasome intracellular protein HLA 5

6 Natural TCRs are present on T-cells and scan other cells for peptide:hla recognition 6 Cancer cells expressing specific cancer antigens will present unique peptides on their HLA Natural T-cell may have a TCR with an affinity for that peptide:hla If affinity and epitope numbers are high enough, an immune synapse will form and T-cell will initiate killing of cancer cell

7 Why do natural T-cells fail to eradicate cancer? TCRs are under positive and negative selection against self antigens Proteins expressed in cancer are self antigens TCRs that bind very strongly or not at all to self antigens are deleted in the thymus Therefore, TCRs that recognise cancer antigens will not be of high affinity Cancer cells often harbour low levels of HLA Complete loss of HLA unlikely to occur NK cells would target such cells The tumour creates an inhibitory and suppressive microenvironment to the immune system Higher load of CD4+ Treg cells, MDSCs, T-cell exhaustion, etc Therefore, in the context of low affinity cancer TCRs, a reduction in epitope number and an inhibitory environment immune tolerance 7

8 What about checkpoint blockade antibodies? Recent success with checkpoint blockade antibodies Targets against CTLA, PD-1, PD-L1 Durability observed however complete responses are low (5-10%) Why? (natural TCR affinity too low, HLA-downregulation) 8

9 Antibodies target cell surface proteins whilst TCRs access both cell surface and intracellular proteins Cancer cell ~10% of cancer targets are cell surface proteins Intracellular target Proteasome Peptides TAP HLA T cell HLA-peptide antigen >90% of cancer targets 9

10 ImmTACs are HLA-peptide targeted bi-specific drugs Cancer cell HLA-peptide Antigen T cell Intracellular protein Proteasome Peptides TAP HLA Immunocore ImmTAC HLA % US EU5 UK Japan China A A A

11 ImmTACs a new class of bi-specific biologic based on the T cell receptor Key points Soluble, highly stable protein (with dominant IP) pm TCR affinity increased several million fold Low target epitope requirements (5-10 per cell) Affinity Clinically proven effector function (anti-cd3) to recruit T-cells 75KDa size provides good tumour penetration Scalable, low cost manufacture (E.coli) nm Early regulatory pathway approved by FDA and MHRA Preliminary clinical data supports MOA, tolerability and efficacy 11 Liddy et al., 2012, Nature Medicine

12 ImmTAC mechanism of action T cell redirection 12

13 ImmTAC mechanism of action T cell redirection 13

14 Visualising ImmTAC-mediated killing Time-lapse microscopy - 11 hour timeframe A375 (HLA-A1 + /MAGE-A3 + ) HLA-A1 + Controls Un-stimulated CD8 + T Cells +0.05nM MAGE specific ImmTAC (HLA-A1 restricted) 14

15 ImmTACs: Empowering T-cells to overcome tumour immunotolerance High affinity TCR to recognise and decorate cancer cell Low pm (5-100pM typical) K D of TCR for peptide:hla Long receptor binding half-life: t 1/2 >12-20h Ability to recruit and redirect host immune T-cells to kill cancer Anti-CD3 scfv molecule with affinity in low nm range Immune synapse forms from decorated cancer cell with multiple anti- CD3 s displayed Evidence that presence of Treg s not inhibitory to potency Evidence of ability to redirect non-classical effector T-cells to kill target cells (e.g. CD4+) Overcoming low epitope display and HLA down-regulation High affinity and optimised anti-cd3 allows killing of cells with as little as 5-10 epitopes displayed 15

16 16 IMCgp100 lead candidate: Preclinical considerations

17 Effector end Targeting end Overview of lead clinical programme - IMCgp100 Target - gp peptide presented by HLA-A2 TCR affinity increased 3,500,000 fold from 85μM to 24 pm K D ~24 pm Residence T ½ ~24 hrs at 37ºC Plasma clearance T ½ ~7 hours in humans K D nm Residence T ½ mins IMCgp100 (HLA-A2) addresses ~48% of western market ~85% market accessible with additional HLAs (abbreviated development pathway) Indications Melanoma and glioma, both treatment and adjunct settings Manufacturing 1,000 vials from 45L run 4 year shelf life so far Regulatory status - CTA/IND approved by MHRA/FDA Clinical status Phase 0 complete Phase I complete (MTD 600ng/Kg) Phase IIa started Q

18 IMCgp100 is human specific on both ends no standard toxicity species available Cancer cell T-cell ImmTAC binds to specific peptide in context of HLA HLA-transgenic mice available, however Point mutations in peptides are common inter-species Point mutations usually abrogate binding of TCR to peptide:hla HLA:peptide antigen E.g. IMCgp100 does not recognise murine gp100 (at homologous sequence) ImmTAC CD3 Anti-CD3 scfv is specific for human CD3 recognition scfv does not bind to standard tox species (including chimp) Transplanted human T-cells are not as active in mice Animals are also usually immunocompromised 18

19 In vitro preclinical approach must be predictive Efficacy: Does the drug kill tumour cells that present the target peptide? Is the drug potent enough to warrant clinical testing? (MABEL) Specificity: Does the drug show low/no recognition of cells that do not present the target peptide? Can we sufficiently cover the peptidome? Does the drug cause any other immune-related adverse events? Cytokine storm, platelet activation, alloreactivity, etc. On-target-off-tumour considerations: gp100 expression: normal melanocytes (skin), retinal cells, inner ear, substantia nigra (is there an acceptable therapeutic window?) Utilise in vitro cellular assays to answer these questions 19

20 Functional in vitro preclinical studies IMCgp100 Effector cells: PBMC/CD8/CD3 T cells Target cells Efficacy: Indication relevant antigen+ tumour cells (HLA appropriate) Specificity/tox: primary human cells (representing range of tissues; HLA appropriate) Readouts: IFNγ (ELIspot), Granzyme B, Cytotoxicity, Caspase activity, Cytokines (including multiplex) 20

21 ImmTAC candidates Clinical Candidate Preclinical Safety Testing and Efficacy Testing: Improved pathway Molecular Analysis Peptide BLAST Mass Spec Alanine and X- Scan Motif BLAST Mass Spec Peptide Screen Peptide Screen Human Cell Testing Normal Primary Cell Screen Platelet Assay Whole Blood Assay Allo-reactivity Screen Differentiated skeletal muscle myoblasts and cardiomyocytes Peptide Screen hit(s) positive cells icells 3D primary cultures Potency/efficacy 2D Cancer cell-lines Peptide family members 3D cancer cell-lines Peptide screen Primary tumours (if available) Cancer Patient T cells (if available) 21 / = addition since IMCgp100 / = expanded upon since IMCgp100

22 Efficacy: IMCgp100 potently redirects T-cells to kill gp100 expressing tumour target cells IMCgp100 redirects T-cells to potently kill HLA- A2+ melanoma targets expressing gp100 Maximal cell killing is observed from hr (donor dependent) Diseased patient cells are equally potent at redirected killing with IMCgp100 22

23 Specificity: IMCgp100 does not significantly react with normal primary cells 9 IMCgp100 does not cross-react with gp100- negative normal primary cells HLA-A*-02 positive human tissue cells Cell origin IFNg Assay GrB Gp100 transcript Summary Melanocytes Expected on target activity observed Hepatocytes X X ND No reactivity Colonic smooth muscle X X NT No reactivity Dermal fibroblasts X X NT No reactivity Pulmonary fibroblasts X X NT No reactivity Bronchial epithelial X X ND No reactivity Bronchial smooth muscle X X NT No reactivity Renal epithelial X X ND No reactivity Astrocytes X X ND No reactivity Dermal MVECs X X ND No reactivity Cardiac myocytes X X ND No reactivity Skeletal muscle X X ND No reactivity Prostate epithelial X X NT No reactivity HLA-A*-02 negative human tissue cells Colonic smooth muscle X X NT No reactivity Pulmonary fibroblasts X NT NT No reactivity Astrocytes X X NT No reactivity Cardiac myocytes X NT NT No reactivity 23 ND indicates not detected and NT indicates not tested

24 Specificity: IMCgp100 does not show signs of cytokine storm, alloreactivity or platelet activation 9 Whole Blood Assay: IMCgp100 does not cause cytokine release in anticipated clinical dose range (<10nM) IMCgp100 does not illicit an alloreactive response when exposed to targets covering >90% of known HLA type IMCgp100 does not cause platelet activation (or interfere with clotting) 24

25 25 IMCgp100: Clinical data

26 IMCgp100 is well tolerated in man Preclinical data package predicted clinical tox observations Expected to observe skin rash with possible vitiligo at higher doses (ontarget) Skin rash observed from Cohort 3 onwards Expected to observe toxicities related to an inflammatory response Pyrexia, oedema, pseudo-lymphopenia observed Dose Limiting Toxicity (DLT) achieved at an expected level of drug DLT was hypotension (likely due to inflammatory response in skin) Preclinical data package predicted efficacy Dosing regime of Phase I not optimal for observing efficacy however some promising signs of efficacy emerged Clinical responses observed including signs of durability Preclinical and Research data paving the way for further optimised Phase IIa dosing regimes 26

27 Example of response after single dose of IMCgp100 Patient dosed once at MTD (600ng/Kg) 30% reduction in diameter 27

28 IMCgp100 shows signs of durability: Patient after 1 year of treatment (at MTD)* Screening Feb 2013 Cycle 2 May 2013 Baseline -26% Cycle 5 Nov 2013 Cycle 6 Feb % -59% 28 *As of July 2014 CT scan, patient response at -72% of baseline

29 Conclusions ImmTACs offer a novel and innovative platform for anti-cancer targeted immunotherapy A wide range of targets are available with TCR-based therapies ImmTACs can harness the power of the immune system by redirecting T-cell activity against tumour cells A sensible in vitro package can be predictive of in vivo observations Not all therapies will be amenable to in vivo preclinical testing The regulatory authorities usually respond well to well thought-out and scientifically grounded strategies of preclinical data Immunotherapies continue to offer an exciting approach towards durable responses in oncology IMCgp100 shows signs of promising efficacy Exciting prospects of the potential to combine with checkpoint inhibitors (CTLA-4, PD-1, PD-L1, etc.) 29

30 Acknowledgements Immunocore: All employees past and present (now too many to list..!) 30